from John Johnston, Machinic Life: The Lure of the Post-Natural (draft) CHAPTER 3 Vibrant Cells: Cellular Automata, Artificial Life, Autopoiesis Yet nature is made better by no mean But nature makes that mean: so, over that art Which you say adds to nature, is an art That nature makes. William Shakespeare, A Winter's Tale Durham tried again, but the Autoverse was holding fast to its laws. Red atoms could not spontaneously appear from nowhere – it would have violated the cellular automaton rules. And if those rules had once been nothing but a few lines of a computer program—a program which could always be halted and rewritten, interrupted and countermanded, subjugated by higher laws—that was no longer true. Zemansky was right: there was no rigid hierarchy of reality and simulation anymore. The chain of cause and effect was a loop now—or a knot of unknown topology. Greg Egan, Permutation City The very idea that machines may begin to reproduce themselves disturbs our conceptual boundaries. What conditions must be filled in order to initiate such a process? In a paper entitled "The General and Logical Theory of Automata," delivered at the Hixon Symposium in 1948, John von Neumann first broached this question from a logico-mathematical point of view. Subsequently, in a series of five lectures delivered at the University of Illinois in 1949, he conceived of a general plan for a self-constructing automaton. Several years later, in "The Theory of Automata: Construction, Reproduction, Homogeneity," a manuscript begun in 1952 but never published during his lifetime, von Neumann worked out an elaborate "schema" that included detailed design criteria and specifications. Even so, it would not be until the late 1970s that Christopher Langton would bring this "schema" to fruition with the creation of what has since become known as Artificial Life. Dovetailing with and relayed by theories of self-organization and emergence in nonlinear dynamical systems being developed concurrently at the Santa Fe Institute, Artificial Life spawned and exfoliated into wondrous forms, fomenting new questions about life and its relation to information. With the birth of these forms a new discipline of research was also inaugurated, reducible to neither theoretical biology nor computer science but drawing fundamentally upon both. 2 As an assemblage producing both new forms of behavior and a new kind of performative discursive practice, Artificial Life challenges cultural boundaries and raises novel philosophical questions. Daniel Dennett, for example, has suggested that Artificial Life is “a new way of doing philosophy,” and not simply a new object for philosophy.1 Like its putative parent discipline, Artificial Intelligence – although the filiation is misleading in fundamental respects-- Artificial Life can be viewed as the creation and testing of thought experiments, with the difference that the experimenter is kept honest by the fact that the experiments are, as Dennett puts it, “prosthetically controlled,” presumably by the “simulational virtuosity of computers.” Although Dennett remains tacitly coy about these important matters, he understands that Artificial Life is no ordinary science, since it creates both a new kind of object and a new methodology. One might say that its object is no longer nature but the simulation of “natural” processes that exhibit life-like behavior, but this formulation hardly conveys the extent to which hard and fast distinctions between the natural and the artificial, phusis and techné, are here undergoing radical revision and transformation.2 For Lacan and Freud before him, the distinction between the realm of the biological and that of the human psychic economy had been a theoretical point d’appui and essential support. At the same time, Lacan's introduction of the symbolic order into human life made it no longer possible to maintain any simple dualism of organism and machine. With Lacan's complex notion of "in-mixing," machines can no longer be said to reside and function outside the realm of human life, as technological extensions or modifications of it; instead, they must be understood to traverse and order it from 1 Daniel Dennett, “Artificial Life as Philosophy,” Artificial Life: An Overview, ed. Christopher Langton (Cambridge, Mass: The MIT Press, 1995), p. 291. 2 One might also say that A-Life is one of the new “sciences of the artificial,’’ to borrow Herbert Simon’s title phrase. In The Sciences of the Artificial (first published in 1969), however, Simon is primarily concerned with questions of rationality and design; indeed, one of his chapters is called “Embedding Artifice in Nature,” and relegates biology and evolutionary process to the sidelines. In contrast, John Holland’s Adaptation in Natural and Artificial Systems (first published in 1975) provides not only groundbreaking conceptual tools but also a much greater foretaste of what is to come. 3 within. In Deleuze and Guattari's sense, these "abstract machines" operate equally in the realm of physical force and mathematical concept, in flows of both matter and information. Yet, while D&G speak of an indeterminism or free play in DNA and the genetic code (a “surplus value of code,” they call it), and hence of a certain becoming- animal and a-parallel evolution, the biological realm remains largely outside their concerns, precisely because of its very capacity to reproduce and self-propagate “naturally,” along lines of filiation. Against this natural mode of propagation, they extoll alliance, monstrous couplings, symbiosis, “side-communication” and “contagion,” and above all those doubly deterritorializing relays they call “becomings.” With the appearance of self-reproducing machines and new forms of machinic "life," however, a whole new branch of the machinic phylum begins to evolve along lines of demarcation and differentiation unlike anything given in nature. In this perspective their marginalizing of biology demands redress.3 Of course, the opposition between organism and machine has never been a simple empirical given; even so, this “molar” dualism has functioned as a mainstay of western metaphysics at least since Aristotle. What has enabled the boundary line between machines and biological organisms or "life" to be drawn and maintained is the basic fact that machines have never been able to reproduce themselves. In Kant's classical formulation a machine possesses only "motive force" [bewegende Kraft], in contrast with a product of nature, which is an "organized being that has within it formative force, and a formative force that this being imparts to the kinds of matter that lack it (thereby organizing them). This force is therefore a formative force that propagates [or reproduces] itself [fortpflanzende bildende Kraft] -- a force that a mere ability [of one thing] to move [another] (i.e., a mechanism) cannot explain."4 What all 3 See Mark Hansen's important essay, "Becoming as Creative Involution?: Contextualizing Deleuze and Guattari's Biophilosophy," in Postmodern Culture 11.1 (Sept. 2000). Although Hansen's purpose is to critique certain excesses in D&G's appropriation of biological theory, he also shows how closely resonant they are with aspects of contemporary biology influenced by complexity theory. 4 Immanual Kant, The Critique of Judgment, trans. Werner S. Pluhar (Indianapolis: Hackett, 1987), p. 253. 4 machines lack (the specific example is a watch) is what Kant calls "natural purpose," which is exhibited by all "organized beings.” Natural purpose, in turn, follows from two conditions, which must both be met: the parts of organized being are produced and exist for each other, and they are all part of a "self-organizing" unity of the whole. Again, machines lack this finality or purposiveness, that is, the self-organizing capacity to be self-directed; they receive not only their formal cause or purpose but also their efficient cause from outside themselves. As Kant states summarily, machines exist only "for the sake of the other." With the advent of machinic self-organization and self-reproduction, these distinctions no longer hold. In fact, as Bernard Stiegler has pointed out, they actually begin to break down with the formation of a “dynamic of the technical system” following the rise of industrialism, and thus implicitly even before self-reproduction becomes an issue.5 For Stiegler, who examines the concept of the “technical system” in the writings of Bertrand Gille, Gilbert Simondon, and André Leroi-Gourhan, what is at stake is the extent to which the biological concept of evolution can be applied to the technical system. In Le mode d'existence des objets techniques (1958), for example, Gilbert Simondon argues that with the Industrial Revolution a new kind of "technical object" distinguished by a quasi-biological dynamic is born. Strongly influenced by cybernetics, Simondon understands this becoming-organic of the technical object as a tendency among the systems and sub-systems that comprise it toward a unity and a constant adaptation to itself and to the changing conditions it brings about. However, when the machines themselves begin to reproduce they attain a dramatic kind of "life" never before imagined, except under the literary or mythic aegis of the demonic and the infernal. Yet it is not from these categories and the thinking they 5 See La technique et le temps, tome 1 (Paris: Galilée, 1994), esp. pp.43-94. Unfortunately Steigler doesn’t consider contemporary developments like Artificial Life. 5 imply that machinic life will liberate